Headscale Tailnet Node Health and Expiry Sentinel
Continuous node connectivity auditing, machine key expiration monitoring, and Slack alerting for Headscale tailnets.
Monitors Traefik Ingress for active connection surges and elevated 5xx error responses to detect ingress saturation.
This workflow connects to the Prometheus metrics port of Traefik Ingress Controller (default port 9100) to audit the traefik_entrypoint_open_connections gauge. Ingress controllers are the entry point for all incoming external customer traffic. During distributed denial-of-service (DDoS) events, flash sales, or connection leaks from misconfigured keepalive clients, open socket connections accumulate, exhausting file descriptor limits (ulimit -n) and worker threads. This workflow detects connection surges early to protect ingress availability.
metrics.prometheus=true).SLACK_WEBHOOK configured in Kestra.traefik_service_requests_total{code=~"5.."} to correlate connection surges with server error rates.